MMLongBench-Doc · Qwen3-VL技术报告v2;Table 2; page 17 text identifies the thinking value as 56.2%; multi-image prompt in Appendix B.4 · accuracy · Appendix B prompts
MMLongBench-Doc · Qwen3-VL技术报告v2;Table 2; page 17 text identifies the thinking value as 56.2%; multi-image prompt in Appendix B.4 · Benchmark definition and scores
MMLongBench-Doc · Qwen3-VL技术报告v2;Table 2; page 17 text identifies the thinking value as 56.2%; multi-image prompt in Appendix B.4 has benchmark disclosures or scores across 1 model release.
Regular display scope. Rankings are grouped by metric and evaluation protocol. Recorded tool access, budgets, and frameworks are shown separately. Scores remain available for reference when conditions are unspecified.
accuracy %
Appendix B prompts
MMLongBench-Doc · Qwen3-VL技术报告v2;Table 2; page 17 text identifies the thinking value as 56.2%; multi-image prompt in Appendix B.4 · accuracy · Appendix B prompts